*h₁es-r-

Based on the Luwian derivatives such as ašraḫi(t)- ‘femininity’ and ašrul(i)- ‘female’, we can reconstruct a non-attested Luwian lexeme *ašr(i)- (and a u-stem *ašru-) with a meaning relating to ‘woman’. A cognate is found in the Hittite suffix -ššara-, which is used to create the feminine counterpart of a noun, e.g. ḫaššu- ‘king’ vs. ḫaššu-ššara- ‘queen’ (HEG S:861-863 with further literature). Regarding the geminate -šš- in the Hittite suffix, Rieken 1999a:262 explains it convincingly (against the previous literature) as representing the assimilation of the initial *h₁ with the root-final -s following a zero-grade of the root reconstructed as *h₁es-.The choice of the laryngeal is left open by HEG S:862, but *h₃ is posited by Yakubovich 2013a:91. The -r- is explained as the remnant of a common gender r-stem, i.e. *h₁s-r-´ > -ššar- (+ thematization) > -ššara-. There is also evidence that the original noun was suffixed in Luwian as well see under CLuw. nanašri(ya)- ‘of the sister’. Since the Luwian derivational bases of none of the derivatives are attested, it is not possible to know exactly what kind of stem we are dealing with. Therefore, no paradigm or stem formation will be proposed here.

[D.S.]

PIE *h₁és-(o)r- is likely to mean ‘woman’ as becomes clear from the morphological analysis presented below. Cf. for the meaning also Watkins 2000a:24 s.v. esōr, Lühr 2000a:142, NIL:681 n. 1.

In Anatolian, formations such as Luw. ašraḫi(t)- ‘femininity’ and ašrul(i)- ‘female’ point to a non-attested Luwian noun *ašr(i)- ‘woman’ vel sim. < *h₁esr-. This is considered to be cognate with the Hittite suffix -ššara-, with which feminine counterpart formations could be formed – cf., e.g., Hitt. ḫaššu- ‘king’ : ḫaššu-ššara- ‘queen’. According to Rieken 1999a:262, -ššara- is likely to go back to *-h₁soro-, a thematized o-grade stem *h₁sor-. The combined evidence of *h₁esr- (whence Luw. ašr° ‘woman’) and *h₁sor- (in Hitt. -ššara-) suggests the reconstruction of an amphikinetic r-stem noun PIE *h₁és-or- ‘woman’. Furthermore, this stem seems to be the second element of the Indo-Iranian and Celtic feminine numerals for ‘3’ and ‘4’: Ved. tisráḥ, Av. tišrō, OIr. téoir, MW teir f. ‘3’ << *tri-h₁sr-és (with laryngeal loss according to the Wetter-rule) and Ved. cátasraḥ, Av. cataŋrō, OIr. cethéoir, MW pedeir f. ‘4’ < *k^wete-sr-es < *k^wétes-h₁sr-es (with *CHCC > *CCC and subsequent simplification of the word-internal geminate -ss-) or < *k^wét-h₁esr-es. For a thorough discussion of the phonological and morphological analysis, cf. Steer in preparation. Another trace of PIE *h₁és-or- might be YAv. hāirišī- woman’: According to Oettinger 1986c:126f., it is based on an iš-stem noun *hā̆r-iš- ‘weibliches Wesen’ < *h₁sō̆r°. Lat. uxōr- is ambiguous and no reliable testimony for a PIE stem *(h₁)sor- ‘woman’. It could continue a preform *h₁uks-sor- ‘housewife’ (formed to *h₁uk-s-, the zero grade of PIE *h₁éu̯k-os/es- n. ‘house’ [> Ved. ókas-] + *sor- ‘woman’, cf. Harðarson 2014a:35). Alternatively, Pinault 2013a:250f. explains the Latin noun from a preform *h₁uk-s-or- ‘the one being in the house’, a delocative formation derived from a loc. sg. *h₁uk-s-ér ‘in the house’. Finally, it has also long been assumed that PIE *h₁és-or- ‘woman’ is contained in PIE *su̯ésor- f. ‘sister’. Perhaps the word for sister is to be analyzed as a “[c]ompound *swe-sor- ‘woman of one’s own (kin group), sister’” (Watkins 2000a:24 s.v. esōr) < PIE *s(u̯)e- + *(h₁)esor- ‘woman’ (cf. Watkins 2000a:89 s.v. swesor-). Similarly Lühr 2000a:141f. There is, however, the competing view that the word for sister is rather to be analyzed as *su̯-ésor- and contains a form of PIE *h₁ésh₂-r/n- ‘blood’ as second element. Cf. NIL:681 n. 1 and Steer in preparation.

[T.S.]

Ved. tisráḥ = Av. tišrō f. ‘3’,← *tri-šr°, cf. Eichner 1982b:202 n. 129, 200 n. 125, EWAia I:677. Ved. cátasraḥ = Av. cataŋrō f. ‘4’;Cf., e.g., EWAia I:675–678, Lühr 2000a:141, NIL:681 n. 1. YAv. hāirišī- f. ‘woman’

?Lat. uxor, gen. uxōris f. ‘wife’ (Naev.+)Cf., e.g., EDLIL:648f., Harðarson 2014a, esp. p. 35). Cf. also Watkins 2000a:24 s.v. esōr.

OIr. téoir, MW teir f. ‘3’, OIr. cethéoir, MW pedeir f. ‘4’Cf., e.g., EWAia I:527, 677f., Lühr 2000a:141, NIL:681 n. 1.

?PIE *su̯ésor- f. ‘sister’ (Ved. svásar-, OIr. siur, Goth. swistar, OCS sestra, etc.)Cf. NIL:680–683.